Add segment groups

[frcroth]

URL of deployed dev instance (used for testing):

• https://segmentgroups.webknossos.xyz

Steps to test:

• create a new annotation
• switch to the segments tab
• create some segments by brushing with some ids or simply click on existing segments
• right click on the root node in the segments tab to create one or multiple groups
• drag segments into the groups
• delete groups
• use undo/redo
• save & reload to check that everything was saved correctly

Todo

• Front end (@philippotto)
  • create groups
  • delete group (recursively)
  • renaming of groups
  • dnd for groups and segments
  • undo/redo for group mutations
  • bugs
    • segments and groups can be moved higher than the root which makes them disappear
    • groups are not shown when no segments exist at all
    • undo might still be buggy. I fixed some bugs, but I still managed to break it at some point. I didn't find a clear reproduction, though.
    • item is moved to wrong group sometimes
  • clean up
• doublecheck export/import via NML (backend does this, for frontend, there is only a skeleton-related download, so no need to do adapt this in my opinion)
• Handle merging of groups
• Fix e2e test

Issues:

• fixes Segments Groups #6954

---

(Please delete unneeded items, merge only when none are left open)

• Updated changelog
• Updated migration guide if applicable
• Updated documentation if applicable
• Adapted wk-libs python client if relevant API parts change
• Removed dev-only changes like prints and application.conf edits
• Considered common edge cases
• Needs datastore update after deployment

[philippotto]

Changing the color of a segment leads to a crash

Oops, this is broken on master currently. I pushed a fix. If I merge this PR on Monday, it should be fine, I think.

[philippotto]

Changing the color of a segment leads to a crash

Oops, this is broken on master currently. I pushed a fix. If I merge this PR on Monday, it should be fine, I think.

@frcroth just pointed out that the back-end review is also pending, so I extracted the fix into #7000 to not create an unnecessary pressure.

[fm3]

Backend LGTM :)

• During testing I noticed that you can get -1 as group id for segments if you drag them into a group and then back into the root group. The -1 should not reach the backend/the NML. I guess this is a front end issue

[philippotto]

The PR should be ready now from my side :) Not sure whether we want to wait for @daniel-wer's final review or maybe it's enough if somebody else does a last round of monkey testing?

Importing an NML with segment groups through the frontend import fails or behaves wrongly, because the segment groups are not distinguished from the tree groups. So either the segment groups are imported as tree groups or the import fails, because of duplicate IDs. See this example NML:
NML_with_segment_groups.zip
I understand that the frontend download doesn't include the segment groups, because it explicitly says "Download selected trees", but maybe it should import the segment groups and segments correctly? Otherwise, the import modal should state that only trees are imported, I think.

The import of skeleton groups should not be confused by segment groups anymore. I started implementing the import of segment groups, but it turned out to be a rabbit hole. In the end, the main problem was that it's not clear into which volume layer the segments should be imported (the id can change). Also, semantics-wise it is weird, because no actual volume data is imported. So, I decided together with @fm3, that the import should stick to skeletons and for everything else, the backend import should be used.

[daniel-wer]

LGTM, well done! I didn't notice any issues during retesting 🎉